The point
A crypto loan lets you borrow rupees or dollars against bitcoin you keep, and the catch is that the lender can sell your bitcoin if its price falls far enough. In the worked example below, a ₹30 lakh loan against ₹1 crore of bitcoin is sold out after a fall of about 57 percent. A ₹50 lakh loan is sold out after about 29 percent. You keep your bitcoin exposure and add a second obligation on top of it.
Can you really borrow against bitcoin without selling it?
Yes, and that is the whole appeal. You pledge bitcoin as collateral (an asset the lender holds as security), receive a loan, and get the bitcoin back when you repay.
The long holder's reasoning is simple. A sale ends the position. In India it also meets the tax regime in the post on crypto tax and the loss set-off you never get. Borrowing looks like a way to raise cash while the position stays intact.
Xapo Bank's chief executive Seamus Rocca described the use case in a 2025 interview. "If you follow the ethos of investing, the smart thing to do would be not to sell it in three days if it goes to $100,000. But life gets in the way." He said Xapo's bitcoin-backed loans come at loan-to-value ratios of 20, 30 and 40 percent.
The same interview contains the risk in one sentence. Rocca said that at a 20 percent ratio, bitcoin would have to fall below $40,000 before a borrower is liquidated. That figure belongs to a conservative loan on one lender's terms, and the rest of this piece shows how quickly it changes with a larger loan.
What is loan-to-value, and why does it decide everything?
Loan-to-value, or LTV, is the loan divided by the market value of the collateral. Borrow ₹30 lakh against ₹1 crore of bitcoin and your LTV is 30 percent.
The bitcoin price moves every day and the loan does not. So LTV moves every day without you doing anything. A fall in bitcoin raises it, and a rise lowers it.
Every lender sets a second, higher line, called the liquidation threshold. Cross it and the lender is allowed to sell your collateral to recover the loan. Aave's own help page describes the mechanism through a health factor: collateral value times the liquidation threshold, divided by the borrowed value. Below 1, the position can be liquidated. Aave's example is $10,000 of ether with an 80 percent threshold and $6,000 borrowed, which gives a health factor of 1.333.
What a margin call and a liquidation actually do
A margin call is the warning. The lender tells you the ratio has moved toward the line and asks you to add collateral or repay part of the loan. Some platforms send no message at all and rely on an automatic sale.
A liquidation is the sale. On Aave, anyone on the network can start a liquidation once a position is eligible. The liquidator receives collateral worth the debt it repays, plus a liquidation bonus. The bonus is paid out of your collateral, so the sale costs you more than the debt it clears.
Aave's page also says up to 50 percent of the debt can be liquidated when the health factor sits above 0.95. At 0.95 or lower, up to 100 percent can be. A fast fall therefore does not give you a polite partial sale.
Worked example: ₹1 crore of bitcoin, three loan sizes
The numbers below are illustrative. They assume a lender that liquidates when the loan reaches 70 percent of the collateral's value. Your lender's line may sit elsewhere, so read the term sheet before you borrow.
At a ₹30 lakh loan, the liquidation value of the collateral is ₹30 lakh divided by 0.70, which is ₹42.86 lakh. Bitcoin would have to fall from ₹1 crore to ₹42.86 lakh, a drop of 57.1 percent.
At a ₹50 lakh loan, it is ₹50 lakh divided by 0.70, which is ₹71.43 lakh. A fall of 28.6 percent triggers the sale.
At a ₹60 lakh loan, it is ₹60 lakh divided by 0.70, which is ₹85.71 lakh. A fall of 14.3 percent triggers it.
Now add interest. Suppose the loan grows by ₹3 lakh over the term, so the ₹30 lakh loan is now ₹33 lakh. The trigger moves to ₹33 lakh divided by 0.70, which is ₹47.14 lakh, and the cushion shrinks from 57.1 percent to 52.9 percent. The ₹3 lakh is an assumption for the arithmetic and not a quoted rate. The direction is the point: the debt only grows while you hold the loan, and the cushion only narrows.
The risks that do not show on the first screen
Four things sit behind the headline ratio.
The first is speed. Bitcoin has fallen a long way inside a few weeks before, as the post on how deep crypto falls went shows. A position that looked safe at a 29 percent cushion can be gone in one bad month.
The second is the timing of the sale. A forced sale happens when the price is low, which is the moment you would least choose to sell. You lose the position at the bottom and still owe any shortfall the sale did not cover.
The third is the lender. The collateral sits with them while the loan runs. The question is the same one a holder asks of any platform: who holds it, in what account, and what does the lender state about reuse? Xapo says it does not lend out client bitcoin, and that is one lender's statement. Ask every lender for theirs in writing.
The fourth is tax. A liquidation is a sale of the asset. For a holder in India, that sale meets the 30 percent tax under section 115BBH and the 1 percent TDS under section 194S. Losses get no set-off or carry forward. Borrowing postpones nothing if the loan goes wrong, because the forced sale can land the same tax on you in a bad month.
What a disciplined holder does instead
My view is that a loan belongs in the plan only when the liquidation line sits far outside any fall you can imagine. The cushion is the number to watch, ahead of the loan amount. A 57 percent cushion and a 14 percent cushion are different products, even though both are called a loan against bitcoin.
Three checks make that concrete before you pledge anything:
- Work out the liquidation value in rupees using the lender's stated threshold, then compare it with how far bitcoin has already fallen in past drawdowns.
- Decide in advance what you will repay or add if the ratio moves halfway to the line, and keep that cash outside the crypto position.
- Ask whether the need is for a few weeks or for years. A short bridge carries far less interest and cushion risk than a long one.
If the real aim is a position that grows without you managing a ratio, the alternative is to build it on a schedule instead of against borrowed money. The post on what a weekly Bitcoin SIP does covers that path. The post on sizing a position to survive a drawdown covers the question a loan forces on you.
Source: Aave Health Factor and Liquidations help page, https://aave.com/help/borrowing/liquidations.
Source: Cointelegraph, Xapo Bank CEO interview on bitcoin-backed loans, published 7 May 2025, https://cointelegraph.com/news/bitcoin-backed-loans-growth-xapo-bank-ceo.
Frequently asked questions
Is a loan against bitcoin a taxable event in India?
Taking the loan is not a sale of the bitcoin. A liquidation is a sale, and a sale of a virtual digital asset meets the 30 percent tax under section 115BBH and 1 percent TDS under section 194S.
What happens if bitcoin falls below the liquidation line?
The lender sells enough collateral to repay the loan, plus any liquidation bonus or fee. On Aave, up to 50 percent of the debt can be cleared when the health factor is above 0.95, and up to 100 percent at 0.95 or lower.
What is a safe loan-to-value ratio for bitcoin?
No ratio is universally safe. Aave's help page says safety depends on the volatility and correlation of the assets. A lower ratio widens the cushion, as the 30 percent loan in the example survives a 57.1 percent fall at a 70 percent liquidation line.
Does interest raise my liquidation risk?
Yes. Interest adds to the debt while the collateral price moves on its own, so the liquidation value rises over the term and the cushion narrows.
Do I still own the bitcoin during the loan?
You keep the economic exposure and get the bitcoin back on repayment, but the lender holds the collateral meanwhile. Ask the lender in writing whether it reuses client collateral.
Crypto investments are subject to market risk. Not financial advice.
“A better allocation begins with a better explanation.”
Qatobit principle
Published construction. Fixed cadence. Versioned control.



